Zirui Gong

I am a first year PhD student in Griffith specializing in AI security, with a strong foundation in both research and teaching. During my first year, I successfully completed three significant research projects while simultaneously taking on teaching responsibilities for three courses each trimester. I also mentored both senior and junior students, guiding their research efforts and helping them develop their academic skills. My experience in managing multiple tasks, coupled with my ability to balance research and teaching roles, has honed my time management skills and demonstrated my capability to contribute effectively in a teaching assistant position.

• Co-first authored "Production Evaluation of Citrus Fruits based on the YOLOv5 compressed by Knowledge Distillation", which has been accepted for publication at the 2023 26th International Conference on Computer Supported Cooperative Work in Design (core B), DOI: 10.1109/CSCWD57460.2023.10152740.
• First authored "AGRAMPLIFIER: Defending Federated Learning Against Poisoning Attacks Through Local Update Amplification", which has been accepted for publication as a regular paper in the IEEE Transactions on Information Forensics & Security (Q1).
• First authored "Not All Edges are Equally Robust: Model Poisoning Attack Against Ranking-Based Federated Learning" (submitted to A* conference).
• Second authored "ConSeg: Contextual Backdoor Attack Against Semantic Segmentation (submitted to A conference).
• Second authored "Fragile Boundaries: Fine-Grained Model Poisoning in Ranking-Based Federated Learning (submitted to A* conference).